NagiosCore Network Monitoring for Alcatel-Lucent Enterprise OmniSwitches

Nagios Core can be used to monitor network devices and servers. Through SNMP and Nagios extensions, detailed information about network status and device status can be obtained. In this article, the implementation of monitoring using Nagios Core will be described based on Alcatel-Lucent Enterprise equipment.

In this case, Nagios Core is installed on an CentOS 7 server with Nagios Core 4.0.8. The IP address 192.168.10.240 is used.

It is recommended to use SNMP to check the device status. SNMPv3 will provide the best security to avoid that information about the network are going to unauthorized users. A user account nagiosadmin is added on the switch to enable that SNMP information can be received by the nagios. On the switch, snmp security pricay-set is choosen. SNMPv3 is necessary to get information. The user is authenticated with the MD5 algorithm. No encryption is used for transmission of SNMP information.

user nagiosadmin password ******** read-only all md5
snmp security privacy-set

In the next step, Nagios Core can be configured.

define host{
	use		generic-switch		
	host_name	OS6860E-P24		; The name we're giving to this switch
	alias		Core-Switch		; Another name associated with the switch
	address		192.168.20.254		; IP address of the switch
	hostgroups	Infrastructure		; Host groups this switch is associated with
	}

In the first step the host has to be defined. A host can be every switch or chassis, which is reachable via an IP-Address. The host has a defined host_name and optionally an alias.

A configuration file for each switch will be implemented. The defined services are

• Ping
• Uptime
• Port Status

Example of Nagios Core network monitoring with OmniSwitch- and OmniAccess-devices.

For Nagios Core ping can be done with the check_ping command.

define service{
	use			generic-service	
	host_name		OS6850E-24		
	service_description	PING		
	check_command		check_ping!200.0,20%!600.0,60%
	normal_check_interval	0.5		
	}

The example shows a check_ping service where a OS6850E-24 is monitored. The warning level is if 20% packet loss or an average ping time over 200 ms occurs. Critical state will be shown if the average ping time is over 600 ms or the packet loss exceeds 60%. The ping interval is 0.5 seconds.

The system uptime gives a good overview when a network device rebooted last time. Especially over night, that might be unrecognized and is visible with Nagios.

define service{
	use			generic-service	
	host_name		OS6850E-24
	service_description	Uptime	
	check_command		check_snmp! -H 192.168.20.254 -o sysUpTime.0 -P 3 -L authNoPriv -U nagiosadmin -a MD5 -A ********
	}

For this command SNMP is used. The host's IP address is used and the SNMP option sysUpTime.0 is checked. AOS returns the system uptime. The other options are necessary to ensure the SNMP communication with OmniSwitches. -L authNoPriv defined authentication but does not enforce encryption. MD5 is used for SNMP as it was defined on the network device before.

The third use case for Nagios Core monitoring of Alcatel-Lucent Enterprise OmniSwitches is the port status monitoring. SNMP is used for this check.

define service{
	use			generic-service
	host_name		OS6850E-24
	service_description	Port 1/23 „Uplink OS6400“ Link Status
	check_command		check_snmp! -H 192.168.20.254 -o 1.3.6.1.2.1.2.2.1.8.1023 -P 3 -L authNoPriv -U nagiosadmin -a MD5 -A ******** -r 1 -m RFC1213-MIB
	notification_period	24x7
	notification_options	w,c,r
	notifications_enabled	1
	contacts		nagiosadmin
	}

The general service settings are the same used for the uptime monitoring. However, the OID tree is used to monitor the port status. The last four digits define the port, which is 1023 - Port 23 in the first switch of a chassis. The alternative is to use the object name. That would be ifOperStatus.1023. How to convert the object name to OID and reverse is explained in this article In addition, notifications are enabled for this check. The contact nagiosadmin is notified 24 hours a day, 7 days a week when the status change to warning or critical and even though the status change to resolved. MIB translations can be found on the dokuwiki and also in the support web.
Furthermore, passwords can be defined in the nagios config file resource.cfg. User variables are used to enable changes of recurring figures. They can be changed in a single point by using this file. In the example, the -A option would be e.g. $USERX$.
Other variables are predefined. The host address can be defined in the host definition. When the host address is needed in services for this host, $HOSTADDRESS$ inserts the IP-Address.